Sputtering coefficients in wall IDS

[hrdoktorx]

The wall IDS contains a structure for sputtering coefficients, very well hidden under the name global_quantities/neutral(i1)/incident_species(i2). It does provide for tabulation of the sputtering coefficients according to the incident species energy, but not the incident angle (while this can make a very large difference). The quantity global_quantities/neutral(i1)/incident_species(i2)/sputtering_physical_coefficient has three dimensions, one of which is for the neutral type (1: cold, 2: thermal, 3: fast), which makes no sense to differentiate if we already have an energy dimension which separates the three populations and could instead be used for an angular dependence.
I would request that an additional global_quantities/neutral(i1)/incident_species(i2)/angles(:) field be added to the structure, and the sputtering coefficients allowed to vary accorgin to the incident angle.
Moreover, the naming convention for the three neutral types must always be understood WITH RESPECT TO THE PLASMA, and NOT the wall. This needs to be made more explicit in the documentation. "Thermal" neutrals are meant to be the ones which temperature has equilibrated with the main plasma ions, but in PWI parlance, would mean neutral emitted at the wall temperature, which in this IDS is referred to as "cold" neutrals. This confusion would be best avoided.

[imbeauf]

These sputtering coefficients were introduced by the ticket IMAS-4287, for the needs of the DYON code (author: Hyun Tae Kim).
I will ask him if he wants to react on this discussion (otherwise we will do the requested changes).

[hyun-tae-kim]

I agree with the comments from @hrdoktorx. It would be useful to add a field for incident ion angles. We requested to add the physical sputtering yield in the wall IDS, and it was to map the output data from DYON. In our case, we don't need it, as DYON assumes a normal incidence of ions as it simulates the plasma initiation phase with an open field configuration. However, the incidence angle data should be useful for other codes such as SOLPS. I also agree with the other comment that 'cold' neutrals should be explicitly stated in the documentation, which is also the case in DYON. We assume that all sputtered neutrals from the wall have the wall temperature.

[imbeauf]

Ok, for the sputtering_physical coefficients the changes are clear.
What should I do for the sputtering_chemical coefficients ? This one also has a dependence on the neutral type (1: cold, 2: thermal, 3: fast). Should I remove it ? (I assume the incident angle doesn't play a role for chemical sputtering)

[hrdoktorx]

By definition, chemical sputtering will emit molecule that are at the wall temperature or nearby. So that distinction seems unneceesary.

[imbeauf]

Ok, so I am removing this dimension for the sputtering_chemical node.
I create a PR

[imbeauf]

Closed after the merge of PR #126